Potent HIV fusion inhibitors againstEnfuvirtide-resistant HIV-1 strainsYuxian He*†‡, Jianwei Cheng§, Hong Lu*, Jingjing Li*, Jie Hu§, Zhi Qi*, Zhonghua Liu†, Shibo Jiang*, and Qiuyun Dai‡§

*Lindsley F. Kimball Research Institute, New York Blood Center, New York, NY 10065; §Institute of Biotechnology, Chinese Academy of Military MedicalSciences, Beijing 100071, China; and †Institute of Pathogen Biology, Chinese Academy of Medical Sciences and Peking Union Medical College,Beijing 100730, China

Edited by Robert C. Gallo, Institute of Human Virology, University of Maryland, Baltimore, MD, and approved September 3, 2008 (received for reviewJuly 29, 2008)

T20 (generic name: Enfuvirtide, brand name: Fuzeon) is the onlyFDA-approved HIV fusion inhibitor that is being used for treatmentof HIV/AIDS patients who have failed to respond to currentantiretroviral drugs. However, it rapidly induces drug resistance invitro and in vivo. On the basis of the structural and functionalinformation of anti-HIV peptides from a previous study, we de-signed an HIV fusion inhibitor named CP32M, a 32-mer syntheticpeptide that is highly effective in inhibiting infection by a widerange of primary HIV-1 isolates from multiple genotypes with R5-or dual-tropic (R5X4) phenotype, including a group O virus (BCF02)that is resistant to T20 and C34 (another anti-HIV peptide). Strik-ingly, CP32M is exceptionally potent (at low picomolar level)against infection by a panel of HIV-1 mutants highly resistant toT20 and C34. These findings suggest that CP32M can be furtherdeveloped as an antiviral therapeutic against multidrug resistantHIV-1.

drug-resistance � gp41 � peptide � six-helix bundle

In the early 1990s, a number of synthetic peptides derived fromthe N- and C-heptad repeat (NHR and CHR) regions of the

HIV-1 envelope glycoprotein (Env) transmembrane subunitgp41 were discovered to have potent anti-HIV activity (1–6).Two of the CHR peptides, C34 and DP-178 (also known as T20),inhibit HIV infection at low nM levels. Biochemical and bio-physical analyses suggest that these CHR peptides inhibit HIV-1Env-mediated membrane fusion by interacting with the viralgp41 NHR region to form heterologous trimer of heterodimerand block gp41 six-helix bundle (6-HB) core formation, a criticalstep in virus–cell fusion (1, 7–9).

T20 (Enfuvirtide, Fuzeon), jointly developed by Trimeris andRoche, was licensed by the US FDA as the first member of a newclass of anti-HIV drugs—HIV fusion inhibitors. Clinical datashow that T20 is effective as a salvage therapy for HIV/AIDSpatients who have failed to respond to current antiretroviraltherapeutics, including reverse transcriptase inhibitors (RTIs)and protease inhibitors (PIs). However, T20 can easily inducedrug resistance, resulting in increasing failure rates in T20-treated patients. Therefore, we sought to develop HIV fusioninhibitors that are effective against T20-resistant HIV.

Here we designed an HIV fusion inhibitory peptide, desig-nated CP32M (Fig. 1), on the basis of findings from our previousstudies on anti-HIV peptides containing a motif (621QIWN-NMT627) located at the upstream region of the gp41 CHR,immediately adjacent to the pocket-binding domain (10), whichis critical for 6-HB formation and stability. Surprisingly, CP32Mis exceptionally potent against T20-resistant HIV-1 strains, withgreat potential to be further developed as an anti-HIV drug fortreatment of HIV/AIDS patients, in particular those unrespon-sive to the first generation HIV fusion inhibitor used in clinics.

ResultsStructure-Based Design of the Anti-HIV Peptide CP32M. We haverecently found that the 621QIWNNMT627 motif located at theupstream region of the gp41 CHR, immediately adjacent to the

pocket-binding domain, is critical for the NHR and CHRinterhelical interactions, and that the peptide CP621–652 con-taining the 621QIWNNMT627 motif possesses potent anti-HIVactivity (10). We designed a peptide, designated CP32M, byusing the peptide CP621–652 as a template, to improve theanti-HIV activity and drug-resistant profiles and the pharma-cokinetics of the anti-HIV peptide with wild-type sequence. Asshown in Fig. 1C, 11 of 32 residues (34.4%) in the peptideCP621–652 were mutated, while the residues that are importantfor the activity or stability of the peptide remained unchanged.The positively or negatively charged residues (e.g., K or E) wereintroduced into the CP32M to promote the formation of ionpairs (salt bridges) at the i to i � 4 position of the helicalconformation (e.g., E636, K640, and K644). First residue Q621 at the‘‘a’’ position in the heptad repeat was replaced by a hydrophobicresidue V to enhance its hydrophobic interaction with the NHRtarget. The residues I622, N625, S640, and N651, which are locatedat the b, c, e, and f positions in the �-helical wheel, were replacedby negatively or positively charged residues, E or K, respectively,to increase the hydrophilicity of the peptide. It was expected thatintroduction of these residues in the CP32M would improve itssolubility and strengthen its ionic interactions with the NHR.

CP32M Is Highly Effective in Blocking HIV-1-Mediated MembraneFusion and Inhibiting Infection by a Broad Spectrum of HIV-1 Isolates.It was important to learn whether the engineered peptideCP32M maintained, or better yet improved its antiviral activity.First, we determined the inhibitory activity of CP32M on HIV-1IIIB-mediated cell–cell fusion by a dye transfer assay. As shownin Fig. 2A, CP32M inhibited cell–cell fusion with an IC50 of 4 nM,approximately sevenfold more potent than T20 (IC50 � 28 nM).Then, we assessed inhibitory activity of CP32M on HIV-1 IIIBinfection. Consistently, CP32M was found to be capable ofinhibiting HIV-1 IIIB infection in MT-2 cells with an IC50 of 5nM (Fig. 2B), approximately fourfold more effective than T20(IC50 � 26 nM). Further, we evaluated the inhibitory activity ofCP32M on infection of peripheral blood mononuclear cells(PBMCs) by a panel of primary group M HIV-1 isolates. Asshown in Table 1, CP32M potently inhibited infection by primaryviruses with distinct genotypes (subtypes A–G) and phenotypes(R5- and R5X4-tropic). Strikingly, CP32M effectively inhibitedinfection by HIV-1 group O virus (BCF02) with an IC50 of 10nM, whereas neither T20 nor C34 exhibited inhibitory activityeven at a concentration as high as 2,000 nM (Fig. 2C). Therefore,CP32M is a potent HIV-1 fusion inhibitor against infection by

Author contributions: Y.H. and Q.D. designed research; J.C., H.L., J.L., J.H., Z.Q., and Z.L.performed research; S.J. contributed new reagents/analytic tools; Y.H. and S.J. analyzeddata; and Y.H. and S.J. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

‡To whom correspondence may be addressed. E-mail: yhe@nybloodcenter.org orqy�dai@yahoo.com.

© 2008 by The National Academy of Sciences of the USA

16332–16337 � PNAS � October 21, 2008 � vol. 105 � no. 42 www.pnas.org�cgi�doi�10.1073�pnas.0807335105

Dow

nloa

ded

by g

uest

on

Mar

ch 8

, 202

0

both laboratory-adapted and primary HIV-1 strains with differ-ent genotypes and phenotypes.

CP32M Is Exceptionally Potent Against T20-Resistant HIV-1 Strains.Subsequently, we sought to determine whether CP32M is ef-fective against HIV-1 strains resistant to T20. A panel ofHIV-1NL4-3 mutants including two T20-sensitive and five T20-resistant strains (11) were used in our experiments. We foundthat double substitutions in the gp41 NHR (V38A/N42D,V38A/N42T, V38E/N42S, and N42T/N43K) also conferredcross-resistance to the peptide C34 (Table 2). Strikingly, CP32Mwas extremely active against both T20- and C34-sensitive and-resistant viruses. CP32M inhibited infection by severalHIV-1NL4-3 mutants (N42S, V38A, and V38A/N42T) at pico-molar (pM) levels. In particular, infection by HIV-1NL4-3 bearingV38A/N42T double mutations was shown to be effectivelyinhibited by CP32M with an IC50 of 2 pM, whereas it was highlyresistant to both T20 and C34.

CP32M Interacts with the gp41 NHR Region to Form Highly Stable�–Helical Complex and Efficiently Blocks the gp41 6-HB Core Forma-tion. We previously showed that the CHR peptide CP621–652can interact with the NHR peptides to form typical 6-HBs (10).

It was of interest to determine whether the engineered peptideCP32M would retain its ability to interact with the NHR peptidesafter substitution of over one-third of its residues. As shown inFig. 3, CP32M could form typical �-helical complexes with theNHR peptides (N36 or T21), similar to the parent peptideCP621–652 as assessed by CD spectra. Impressively, thermaldenaturation analysis demonstrated that CP32M also interactedwith N36 or T21 with higher thermostability (Fig. 3 B and D).

Fig. 1. Interactions of CP32M and other CHR peptides with NHR peptides. (A)Schematic view of the HIV-1HXB2 gp41 molecule. FP, fusion peptide; NHR,N-terminal heptad repeat; CHR, C-terminal heptad repeat; TR, tryptophan-rich domain; TM, transmembrane domain; CP, cytoplasmic domain. (B) In thecurrent fusion model, the CHR region of gp41 folds back to interact with theNHR region to form a hairpin structure. Three molecules of hairpins associatewith each other to form a 6-HB. The dashed lines between the NHR and CHRregions indicate the interaction between the residues located at the e, g andthe a, d positions in the NHR and CHR, respectively. The interaction of thepocket-binding domain in the CHR (amino acids 628–635, in blue) with thepocket-forming domain in the NHR (amino acids 565–581, in red) is critical forstabilization of the 6-HB. Both T20 and C34 peptides contain the sequencestargeting the GIV motif (in purple) in the NHR, but C34 contains the pocket-binding sequence, while T20 does not. The peptide CP621–652 overlaps thepocket-binding domain but it has no sequence targeting the GIV motif. Theresidues preceding the pocket-binding domain are in green. (C) Based onthe CP621–652 sequence, CP32M was engineered by replacing 11 of 32residues (the mutated residues are in red) to improve its pharmacokineticprofiles and anti-HIV activity. The positively charged residues (e.g., K) ornegatively charged residues (e.g., E) were introduced to form ion pairs (saltbridges) at i to i � 4 position of the helical conformation.

Fig. 2. Inhibition of CP32M on HIV-1-mediated cell–cell fusion (A), and oninfection by HIV-1 IIIB (B) and BCF02 (C). T20 was tested as a control. Eachsample was tested in triplicate and the data are presented as mean � standarddeviations. Numbers in parentheses indicate IC50 (nM) values.

He et al. PNAS � October 21, 2008 � vol. 105 � no. 42 � 16333

MIC

ROBI

OLO

GY

Dow

nloa

ded

by g

uest

on

Mar

ch 8

, 202

0

The complex of N36/CP32M had a Tm value of 79°C which was15°C higher than its wild-type bundle N36/CP621–652 (Tm �64°C). The T21/CP32M complex displayed a Tm value (94°C)13°C higher than the T21/CP621–652 complex (Tm � 81°C). Asa control, the bundle N36/C34, which has been considered to bea core structure of the fusion-active gp41, had a Tm value of 64°C(data not shown).

We then used an N-PAGE-based method to visualize thecomplex formed by CP32M and counterpart peptide T21. Asshown in Fig. 4A, the NHR peptide T21 shows no band in thenative gel because it carries net positive charges and couldmigrate up and off the gel, but the negatively charged peptideCP32M shows a specific band. When CP32M was mixed withT21, a specific band corresponding to the 6-HB appeared. Theirspecific binding was also confirmed by size-exclusion high-performance liquid chromatography (HPLC) (Fig. 4B). Sedi-mentation equilibrium ultracentrifugation demonstrated thatthe MWapp value of the CP32M and T21 complex was 26,600 Da(Fig. 4C). Compared to an expected molecular mass of 8,688 Dafor CP32M/T21 heterodimer, we concluded that these twopeptides associate to form a 6-HB structure consisting of threeCP32M and T21 peptides, respectively.

The mechanism of NHR or CHR-derived anti-HIV peptideshas been considered to inhibit the formation of the viral gp416-HB in a dominant-negative fashion (7, 12). To test whether theengineered CP32M could arrest the formation of 6-HBs, anELISA-based method was developed, in which the 6-HB-specificmAb NC-1 was used as a capture antibody and the peptide C34was biotinylated (see Materials and Methods). Consistently, NC-1reacted specifically with the complex of N36 and C34 but notwith the isolated peptides (data not shown). The results showthat CP32M could efficiently inhibit the formation of 6-HBbetween the peptides N36 and C34-biotin in a dose-dependentmanner, comparable to its parent peptide CP621–652 and C34

itself (Fig. 4D). However, T20 had no such effect at a concen-tration as high as 8,000 nM, consistent with our previous data(13). This result suggests that CP32M, unlike T20, is able to block6-HB formation between the NHR and CHR in a dominant-negative fashion.

DiscussionIn the present study, we designed an anti-HIV peptide, CP32M,on the basis of the structural and functional information ofHIV-1 gp41 and a recently identified anti-HIV peptide (CP621–652) containing a motif (621QIWNNMT627) that is critical for the6-HB formation and stability (10). Our data have demonstratedthat, like its parent peptide CP621–652 (10), the engineeredCP32M maintains its potency in inhibiting HIV-1-mediatedcell–cell fusion and infection by laboratory-adapted HIV-1strains. CP32M is highly effective against a panel of primaryHIV-1 strains with distinct genotypes (group M, subtypes A–G)and phenotypes (R5 and R5X4) (Table 1). Favorably, CP32Mcould potently inhibit infection by BCF02, one of the HIV-1group O isolates, having a high genetic diversity compared to themajor group of HIV-1. In comparison, T20 and C34 had noinhibitory activity against infection by HIV-1BCF02 at a concen-tration as high as 2,000 nM. Therefore, the engineered peptideCP32M which has a broader anti-HIV spectrum than T20 maypossess a property to overcome the genetic barrier of HIV-1group O isolates (e.g., BCF02) and thus can be used fortreatment of HIV-1 group O infection.

The most unique feature of the CP32M peptide is its excep-tional potency against HIV-1 variants resistant to T20 and otherCHR peptides including C34 and T1249, a second generationHIV fusion inhibitor. Although its parent peptide (CP621–652)is also effective against the drug-resistant viruses, CP32M ex-hibited much-improved antiviral activity against some T20-resistant mutants, with IC50 in the picomolar range.

Why are CP32M and its parent peptide CP621–652 effectiveagainst HIV-1 variants resistant to T20, C34, and T1249? Webelieve that this is because they have different target sites in thegp41 NHR region. Early in vitro studies indicate that HIV-1acquires T20 resistance by mutations in the ‘‘GIV’’ motif (po-sitions 36–38 based on reference HIV-1HXB2 gp41 numbering,underlined in Fig. 1B) of the gp41 NHR region (11). Subsequentclinical data show that HIV-1 isolates from patients failingtherapy with T20 also contain mutations in the NHR region ofgp41 (amino acids 36–45: GIVQQQNNLL) (14, 15). Althoughtwo changes within the amino acids 36–45 domain have beenobserved in some patients resistant to T20 therapy, in most casesa single mutation alone can mediate resistance (16, 17). Thesefindings suggest that the GIV motif may be a critical binding sitefor T20. Indeed, Chang and colleagues have shown that theLLSGIV stretch in NHR is a crucial docking site for T20 (18, 19).

Our previous studies demonstrated that the CHR region ofHIV-1 gp41 contains three functional domains (20): a pocket-

Table 1. Inhibitory activity of CP32M against infection byprimary HIV-1 isolates

Virus Subtype Coreceptor

nM � SD

IC50 IC90

RW92008 A R5 51 � 3 66 � 394UG103 A R5X4 30 � 12 54 � 1492US657 B R5 235 � 42 423 � 40JR-FL B R5 23 � 3 37 � 493IN101 C R5 37 � 2 80 � 593MW959 C R5 32 � 3 83 � 1092UG001 D R5X4 40 � 6 87 � 892THA009 E R5 94 � 16 120 � 3093BR020 F R5X4 23 � 3 105 � 7RU570 G R5 84 � 12 132 � 30

Table 2. Potent inhibitory activity of CP32M against infection by C34 and T-20-resistantviruses

HIV-1NL4-3 (36G) Phenotype*

IC50 � SD, nM

T20 C34 CP32M

Parental S 35.820 � 14.350 3.410 � 0.300 4.250 � 0.130N42S S 28.915 � 0.881 0.611 � 0.142 0.042 � 0.004V38A R �2,000.000 1.686 � 0.620 0.220 � 0.075V38E/N42S R �2,000.000 79.660 � 4.630 4.690 � 0.630N42T/N43K R �2,000.000 119.681 � 26.160 1.016 � 0.359V38A/N42T R �2,000.000 50.049 � 8.305 0.002 � 0.000V38A/N42D R �2,000.000 24.474 � 9.246 76.251 � 11.277

*S, sensitive; R, resistant.

16334 � www.pnas.org�cgi�doi�10.1073�pnas.0807335105 He et al.

Dow

nloa

ded

by g

uest

on

Mar

ch 8

, 202

0

binding domain (amino acids 628–635, blue in Fig. 1B) that canbind to the pocket-forming region in NHR (red in Fig. 1B), anHR-binding domain (amino acids 628–666) which is able tointeract with the 4–3 heptad repeat (HR) sequences in the gp41NHR, and a tryptophan-rich lipid-binding domain (amino acids666–673, orange in Fig. 1B) that has a tendency to bind lipidmembranes (20). The helical packing interactions between theNHR and CHR play an essential role in viral infectivity (21–23).In particular, the hydrophobic and ionic interactions of the deeppocket region in the C terminus of the NHR groove andpocket-binding residues from the CHR can determine theconformation and stability of the fusion-active gp41 6-HBstructure (1, 19, 24). T20 contains the HR- and lipid-bindingdomains. It inhibits HIV fusion by binding to the HR sequencesin NHR, including the GIV motif, through its N-terminalHR-binding domain and interacting with the lipid membrane onthe target cell, via its C-terminal lipid-binding domain (25).Mutations of the conserved GIV motif may affect the binding ofT20 to the HR sequence in NHR, resulting in significantreduction of T20-mediated inhibitory activity on HIV fusion andentry. C34 contains the pocket-binding domain and shares withT20 the HR-binding sequence and thus associates with NHR toform a 6-HB through its interaction with both pocket-formingand HR-sequences in NHR. The pocket-forming region and theLLSGIV stretch in NHR are critical docking sites for C34 (18,19). Thus, mutation of the GIV motif in viral gp41 may alsoaffect C34 binding to the HR sequence, leading to resistance toC34. However, binding of C34 to the hydrophobic pocket regionmay partially compensate the decreased binding of C34 to NHR.Therefore, the viruses with GIV mutation in gp41 may be lessresistant to C34 than T20 (26). T1249, a second generation HIVfusion inhibitor, contains all three functional domains, includingpocket-, HR- and lipid-binding sequences. However, it functionsmore like T20 (25). Therefore, T20-resistant strains with GIVmutations are also insensitive to T1249 (27).

Unlike T20 and C34, CP621–652 does not contain the GIV-binding sequence, but consists of the pocket-binding domain andthe 621QIWNNMT627 motif, which is located at the upstreamregion of the CHR and immediately adjacent to the pocket-binding domain and is highly important for the stabilization ofthe gp41 core structure (10). Therefore, the mutations of theGIV motif may have little or no effect at all on the interactionof CP621–652 with the viral gp41 NHR region and consequentlyon the effectivity of CP621–652 against T20-resistant HIV-1strains.

Like its parent peptide CP621–652, CP32M contains noGIV-binding sequence (Fig. 1 B and C). It is expected to beefficient in inhibiting infection by T20-resistant HIV-1 variantswith GIV mutations in the gp41 NHR region. After optimizationof the CP621–652 sequence, the engineered CP32M demon-strated improved anti-HIV activity. Biophysical characterizationshowed that CP32M could form highly stable 6-HBs with thecounterpart NHR peptide T21 that contains no GIV motif, andhad a Tm value of 94°C, while the 6-HB formed by CP621–652and T21 had a Tm of 81°C (Fig. 3D). This result suggests thatCP32M may target the NHR with higher affinity than CP621–652. This may also explain why CP32M is much more potent thanCP621–652 in inhibiting infection by HIV-1 strains resistant toT20, C34, and T1249. All these results suggest that CP32M,which has a shorter peptide sequence than T20 (36-mer) andT1249 (39-mer), has great potential to be further developed asa unique anti-HIV drug for treatment of HIV/AIDS patientswho have failed to respond to the first and second generationHIV fusion inhibitors.

Materials and MethodsPeptide Synthesis. A set of peptides derived from the NHR (N36 and T21) orCHR (CP621–652, C34, and T20) of HIV-1 gp41 and CP32M (Fig. 1) weresynthesized by a standard solid-phase FMOC method using an Applied Bio-systems model 433A peptide synthesizer. All peptides were acetylated at theN termini and amidated at the C termini. The peptides were purified to

Fig. 3. Biophysical characterization of CP32M by CD spectroscopy. (A) �-helical conformation of the complex formed by N36 and CP32M or CP621–652. (B)Thermostability of the complex formed by N36 and C peptides. The unfolding temperature of each complex was scanned at 222 nm by CD spectroscopy, and theirTm values were calculated. (C) �-helical conformation of the complex formed by T21 and CP32M or CP621–652. (D) Thermostability of the complex formed byT21 and C peptides. Final concentration of each peptide in PBS is 10 �M.

He et al. PNAS � October 21, 2008 � vol. 105 � no. 42 � 16335

MIC

ROBI

OLO

GY

Dow

nloa

ded

by g

uest

on

Mar

ch 8

, 202

0

homogeneity (�95% purity) by HPLC and identified by laser desorption massspectrometry (PerSeptive Biosystems, Framingham, MA). The concentration ofpeptides was determined by UV absorbance and a theoretically calculatedmolar-extinction coefficient � (280 nm) of 5500 mol/L�1�cm�1 and 1490 mol/L�1�cm�1 based on the number of tryptophan (Trp) residues and tyrosine (Tyr)residues (all of the peptides tested contain Trp and/or Tyr), respectively.

Circular Dichroism (CD) Spectroscopy. CD spectroscopy was performed aspreviously described (24). Briefly, an N peptide was incubated with a C peptide(10 �M) at 37°C for 30 min. The CD spectra of the isolated peptides and theirmixtures were acquired on Jasco spectropolarimeter (Model J-715, Jasco Inc.,Japan). The �-helical content was calculated from the CD signal by dividing themean residue ellipticity at 222 nm by the value expected for 100% helixformation (i.e., 33,000° cm2 dmol�1) according to the previous studies (28, 29).Thermal denaturation was monitored at 222 nm by applying a thermalgradient of 2°C/min in the range of 4–98°C. The melting curve was smooth-ened, and the midpoint of the thermal unfolding transition (Tm) values wascalculated using Jasco software utilities as described previously (30).

Native Polyacrylamide Gel Electrophoresis (N-PAGE) Assay. N-PAGE was carriedout to determine the 6-HB formation between the N and C peptides asdescribed previously (31). Briefly, N peptide T21 was mixed with C peptideCP32M (40 �M) and was loaded onto a 10 � 1.0-cm precast 18% Tris-glycinegel (Invitrogen, Carlsbad, CA). Gel electrophoresis was carried out with 125 Vconstant voltage at room temperature for 2 h. The gel was then stained withCoomassie Blue and imaged with a FluorChem 8800 Imaging System (AlphaInnotech).

Binding Assays by Size-Exclusion Chromatography (32). T21 was mixed withCP32M (final concentration � 0.20 mM) at a molar ratio of 1:1 in 50 mM

sodium phosphate/150 mM NaCl (pH 7.2) and incubated at 37°C for 30 min.The mixture or peptide (30 �l) was applied to the TSK-G 3000SWxl HPLC columnequilibrated with 50 mM sodium phosphate/150 mM NaCl and eluted at 0.8ml/min, and fractions were monitored at 214 nm.

Sedimentation Equilibrium Centrifugation. Sedimentation equilibrium exper-iments were performed using an Optima XL-I analytical ultracentrifuge (Beck-man) equipped with a standard two-channel cell in an An-60 Ti rotor (33). Thedesignated peptide concentration was 25 �M in buffer consisting of 50 mMsodium phosphate/100 mM NaCl (pH 7.4), and the complex was composed of12.5 �M N peptide (T21) and 12.5 �M C peptide (CP32M). The samples wererun at 25,000 or 33,000 rpm at 20°C for 24 h. Absorbance monitoring wasperformed at 280 nm. The apparent molecular weight (MWapp) was obtainedby fitting the data to self-association using the sedimentation analysis soft-ware supplied by Beckman. The partial specific volumes used for T21 andCP32M were 0.738 and 0.729 respectively, as calculated from the mass averageof the partial specific volumes of the individual amino acids.

Inhibition of CP32M on 6-HB Formation. Inhibitory activity of the peptides(CP32M, CP621–652, C34, and T20) on the 6-HB formation was measured by amodified ELISA-based method as previously described (10). Briefly, a 96-wellpolystyrene plate was coated with a 6-HB specific monoclonal antibody NC-1IgG (4 �g/ml in 0.1 M Tris, pH 8.8). A test peptide at graded concentrations wasmixed with C34-biotin (0.25 �M) and incubated with N36 (0.25 �M) at roomtemperature for 30 min. The mixture was then added to the NC-1-coatedplate, followed by incubation at room temperature for 30 min and washingwith a washing buffer (PBS containing 0.1% Tween 20) three times. Then,streptavidin-labeled horseradish peroxidase (Invitrogen) and the substrate3,3�,5,5�-tetramethylbenzidine (Sigma) were added sequentially. Absorbance

Fig. 4. Determination of the activity of CP32M to form 6-HB with T21 and to block 6-HB formation between N36 and C34. (A) N-PAGE for detection of 6-HBformation between T21 and CP32M. (B) Size-exclusion HPLC analysis for 6-HB formation between T21 and CP32M. (C) Molecular mass of the T21/CP32M complexdetermined by sedimentation equilibrium ultracentrifugation at concentrations of 25 �M in PBS buffer (pH 7.4) at a rotor speed of 33,000 rpm. The observedmolecular mass is 26,600 Da (the calculated mass for a trimer is 26,064 Da). (D) Inhibition of C peptides on formation of 6-HB modeled by N36/C34 peptides.

16336 � www.pnas.org�cgi�doi�10.1073�pnas.0807335105 He et al.

Dow

nloa

ded

by g

uest

on

Mar

ch 8

, 202

0

at 450 nm (A450) was measured using an ELISA reader (Ultra 384, Tecan). Thepercentage of inhibition by the peptides and the IC50 values were calculatedas previously described (34).

Cell–Cell Fusion Assay. A dye transfer assay was used for detection of HIV-1-mediated cell–cell fusion as previously described (35). Briefly, Calcein-AM-labeled H9/HIV-1IIIB-infected cells were incubated with MT-2 cells (ratio � 1:5)at 37°C for 2 h in the presence or absence of the test peptide. The fused andunfused calcein-labeled HIV-1-infected cells were counted under an invertedfluorescence microscope (Zeiss) with an eyepiece micrometer disk. The per-centage of inhibition of cell–cell fusion and the IC50 values were calculated asdescribed before (35).

Inhibition of HIV-1IIIB and T20-Resistant Virus. The inhibitory activity of CP32M,T20, or C34 on infection by various T20-resistant virus isolates and laboratory-adapted HIV-1 strain (HIV-1IIIB) was determined as previously described (35). Inbrief, 1 � 104 MT-2 cells were infected with HIV-1 isolates at 100 TCID50 (50%tissue culture infective dose) in 200 �l culture medium in the presence orabsence of the test peptide overnight. Then the culture supernatants wereremoved and fresh media were added. On the fourth day postinfection, 100

�l of culture supernatants were collected from each well, mixed with equalvolumes of 5% Triton X-100, and assayed for p24 antigen by ELISA.

Inhibition of CP32M on Primary Viruses. The inhibitory activity of CP32Magainst a panel of primary HIV-1 isolates was determined as previously de-scribed (35). Briefly, the PBMCs were isolated from the blood of healthy donorsusing a standard density gradient (Histopaque-1077, Sigma) centrifugation.After incubation at 37°C for 2 h, the nonadherent cells were collected andresuspended at 5 � 105/ml in RPMI medium 1640 containing 10% FBS, 5 �g ofphytohemagglutinin (PHA)/ml, and 100 U of interleukin-2/ml, followed byincubation at 37°C for 3 days. The PHA-stimulated cells were infected with thecorresponding primary HIV-1 isolates at a multiplicity of infection (MOI) of0.01 in the absence or presence of CP32M at graded concentrations. Thesupernatants were collected 7 days postinfection and tested for p24 antigenby ELISA.

ACKNOWLEDGMENTS. We thank Ms. Veronica Kuhlemann for editorial as-sistance. This work was supported by the 973 Program (Grant 2006CB504200)and 863 Program (Grant 2006A09Z404) from the Chinese Ministry of Scienceand Technology, the Nature Science Foundation of China (Grant 30870123),and the National Institutes of Health (Grant AI46221).

1. Chan DC, Chutkowski CT, Kim PS (1998) Evidence that a prominent cavity in the coiledcoil of HIV type 1 gp41 is an attractive drug target. Proc Natl Acad Sci USA 95:15613–15617.

2. Jiang S, Lin K, Strick N, Neurath AR (1993) HIV-1 inhibition by a peptide. Nature 365:113.3. Lu M, Blacklow SC, Kim PS (1995) A trimeric structural domain of the HIV-1 transmem-

brane glycoprotein. Nat Struct Biol 2:1075–1082.4. Wild C, Oas T, McDanal C, Bolognesi D, Matthews T (1992) A synthetic peptide inhibitor

of human immunodeficiency virus replication: Correlation between solution structureand viral inhibition. Proc Natl Acad Sci USA 89:10537–10541.

5. Wild C, Greenwell T, Matthews T (1993) A synthetic peptide from HIV-1 gp41 is a potentinhibitor of virus-mediated cell-cell fusion. AIDS Res Hum Retroviruses 9:1051–1053.

6. Wild CT, Shugars DC, Greenwell TK, McDanal CB, Matthews TJ (1994) Peptides corre-sponding to a predictive alpha-helical domain of human immunodeficiency virus type1 gp41 are potent inhibitors of virus infection. Proc Natl Acad Sci USA 91:9770–9774.

7. Chan DC, Kim PS (1998) HIV entry and its inhibition. Cell 93:681–684.8. Liu S, Wu S, Jiang S (2007) HIV entry inhibitors targeting gp41: From polypeptides to

small-molecule compounds. Curr Pharm Des 13:143–162.9. Roux KH, Taylor KA (2007) AIDS virus envelope spike structure. Curr Opin Struct Biol

17:244–252.10. He Y, et al. (2008) Identification of a critical motif for the human immunodeficiency

virus type 1 (HIV-1) gp41 core structure: Implications for designing novel anti-HIVfusion inhibitors. J Virol 82:6349–6358.

11. Rimsky LT, Shugars DC, Matthews TJ (1998) Determinants of human immunodeficiencyvirus type 1 resistance to gp41-derived inhibitory peptides. J Virol 72:986–993.

12. Weiss CD (2003) HIV-1 gp41: Mediator of fusion and target for inhibition. AIDS Rev5:214–221.

13. He Y, et al. (2008) Design and evaluation of Sifuvirtide, a novel HIV-1 fusion inhibitor.J Biol Chem 283:11126–11134.

14. Mink M, et al. (2005) Impact of human immunodeficiency virus type 1 gp41 amino acidsubstitutions selected during enfuvirtide treatment on gp41 binding and antiviralpotency of enfuvirtide in vitro. J Virol 79:12447–12454.

15. Wei X, et al. (2002) Emergence of resistant human immunodeficiency virus type 1 inpatients receiving fusion inhibitor (T-20) monotherapy. Antimicrob Agents Chemother46:1896–1905.

16. Greenberg ML, Cammack N (2004) Resistance to enfuvirtide, the first HIV fusioninhibitor. J Antimicrob Chemother 54:333–340.

17. Poveda E, Briz V, Soriano V (2005) Enfuvirtide, the first fusion inhibitor to treat HIVinfection. AIDS Rev 7:139–147.

18. Chang DK, Cheng SF, Trivedi VD (1999) Biophysical characterization of the structure ofthe amino-terminal region of gp41 of HIV-1. Implications on viral fusion mechanism.J Biol Chem 274:5299–5309.

19. Chang DK, Hsu CS (2007) Biophysical evidence of two docking sites of the carboxylheptad repeat region within the amino heptad repeat region of gp41 of humanimmunodeficiency virus type 1. Antiviral Res 74:51–58.

20. Liu S, et al. (2007) HIV gp41 C-terminal heptad repeat contains multifunctionaldomains. Relation to mechanisms of action of anti-HIV peptides. J Biol Chem 282:9612–9620.

21. Follis KE, Larson SJ, Lu M, Nunberg JH (2002) Genetic evidence that interhelical packinginteractions in the gp41 core are critical for transition of the human immunodeficiencyvirus type 1 envelope glycoprotein to the fusion-active state. J Virol 76:7356–7362.

22. Liu J, Wang S, Hoxie JA, LaBranche CC, Lu M (2002) Mutations that destabilize the gp41core are determinants for stabilizing the simian immunodeficiency virus-CPmac enve-lope glycoprotein complex. J Biol Chem 277:12891–12900.

23. Lu M, et al. (2001) Structural and functional analysis of interhelical interactions in thehuman immunodeficiency virus type 1 gp41 envelope glycoprotein by alanine-scanning mutagenesis. J Virol 75:11146–11156.

24. He Y, et al. (2007) Conserved residue Lys574 in the cavity of HIV-1 Gp41 coiled-coildomain is critical for six-helix bundle stability and virus entry. J Biol Chem 282:25631–25639.

25. Eggink D, et al. (2008) Selection of T1249-resistant human immunodeficiency virus type1 variants. J Virol 82:6678–6688.

26. Armand-Ugon M, Gutierrez A, Clotet B, Este JA (2003) HIV-1 resistance to the gp41-dependent fusion inhibitor C-34. Antiviral Res 59:137–142.

27. Chinnadurai R, Rajan D, Munch J, Kirchhoff F (2007) Human immunodeficiency virustype 1 variants resistant to first- and second-version fusion inhibitors and cytopathic inex vivo human lymphoid tissue. J Virol 81:6563–6572.

28. Shu W, et al. (2000) Helical interactions in the HIV-1 gp41 core reveal structural basisfor the inhibitory activity of gp41 peptides. Biochemistry 39:1634–1642.

29. Chen YH, Yang JT, Chau KH (1974) Determination of the helix and beta form of proteinsin aqueous solution by circular dichroism. Biochemistry 13:3350–3359.

30. Liu S, et al. (2005) Different from the HIV fusion inhibitor C34, the anti-HIV drug Fuzeon(T-20) inhibits HIV-1 entry by targeting multiple sites in gp41 and gp120. J Biol Chem280:11259–11273.

31. Liu S, Zhao Q, Jiang S (2003) Determination of the HIV-1 gp41 fusogenic core confor-mation modeled by synthetic peptides: Applicable for identification of HIV-1 fusioninhibitors. Peptides 24:1303–1313.

32. Dai Q, Zajicek J, Castellino FJ, Prorok M (2003) Binding and orientation of conantokinsin PL vesicles and aligned PL multilayers. Biochemistry 42:12511–12521.

33. Dai Q, Castellino FJ, Prorok M (2004) A single amino acid replacement results in theCa2�-induced self-assembly of a helical conantokin-based peptide. Biochemistry43:13225–13232.

34. Jiang SB, Lin K, Neurath AR (1991) Enhancement of human immunodeficiency virustype 1 infection by antisera to peptides from the envelope glycoproteins gp120/gp41.J Exp Med 174:1557–1563.

35. Jiang S, et al. (2004) N-substituted pyrrole derivatives as novel human immunodefi-ciency virus type 1 entry inhibitors that interfere with the gp41 six-helix bundleformation and block virus fusion. Antimicrob Agents Chemother 48:4349–4359.

He et al. PNAS � October 21, 2008 � vol. 105 � no. 42 � 16337

MIC

ROBI

OLO

GY

Dow

nloa

ded

by g

uest

on

Mar

ch 8

, 202

0